If lungs could be retrieved from cadavers after circulatory arrest, the critical shortage of donors for
lung transplantation might be alleviated. To assess gas exchange after
transplantation of lungs from cadaveric donors, we performed double-
lung transplantation through sequential
thoracotomies in 12 dogs. Donors were sacrificed by intravenous
pentobarbital injection and then ventilated with 100%
oxygen. Lungs were harvested 2 hours (n = 6) or 4 hours (n = 6) after death and flushed with 2 L modified
Euro-Collins solution. Recipients underwent sequential right and left
lung transplantation; they were then monitored while under
anesthesia for 8 hours, with adjustments of the fraction of inspired
oxygen. Nine of 12 recipients survived the 8-hour study period. Four of six dogs with cadaveric lungs retrieved 2 hours after death survived; deaths were from
pulmonary embolism at 6 hours and
pulmonary edema at 2 hours. Five of six dogs with cadaveric lungs retrieved 4 hours after death survived; one died of
hypoxia during implantation of the left lung, while dependent on the right lung graft. Postoperative hemodynamic and gas exchange parameters were similar in both groups. Alveolar-arterial
oxygen gradient rose significantly compared with baseline 1 hour after
transplantation in both groups (462 +/- 60 vs 38 +/- 31 mmHg for 2-hour group, p < 0.0001, and 484 +/- 63 vs 38 +/- 14 mmHg for 4-hour group, p < 0.0002). By 8 hours after operation, the gradients had significantly decreased in both groups (105 +/- 37 mm Hg for 2-hour group and 146 +/- 53 mm Hg for 4-hour group) and were similar to baseline values. Extravascular lung water also rose significantly 1 hour after
transplantation (15.7 +/- 2.8 vs 7.9 +/- 0.5 ml/kg for 2-hour group, p < 0.02, and 16.9 +/- 1.2 vs 6.6 +/- 0.4 ml/kg for 4-hour group, p < 0.0001) and decreased gradually during the 8-hour study period. Donor lungs retrieved at 2 and 4 hours postmortem afford similar recipient outcomes. Improvement in alveolar-arterial
oxygen gradient and reduction in extravascular lung water during the study period imply that the
ischemia-reperfusion injury induced by this model is reversible. If this approach could be safely introduced to clinical practice, substantially more transplant procedures could be performed.